The United States is preparing to tell dozens of countries they must pick sides in the artificial intelligence race with China, warning they will be excluded from a US-led AI coalition if they also partner with Beijing, Reuters reported on August 14, 2026, in an exclusive by Washington correspondent Michael Martina.
The warning comes in a draft letter prepared by the State Department and addressed to the 35 signatories of a US "AI Opportunity Statement" signed in June, a group that includes members of the non-binding Pax Silica framework, according to the report. The move marks a sharp escalation in the technological Cold War between Washington and Beijing, and AI Buzz Wire is following the developing story as details of the draft continue to emerge. For more context on this story, see our ongoing AI industry coverage.
What the Draft Letter Says
According to the leaked draft obtained by Reuters, the letter pressures signatories to treat their participation in US-led AI arrangements as an exclusive commitment. "Signature of the Pax Silica Declaration is not merely a membership subscription, but a commitment," the draft states, urging countries to "choose deliberately" about their technology partnerships.
The implicit threat is exclusion: countries that also sign up for Chinese AI initiatives or infrastructure partnerships risk being locked out of the US-led coalition and the benefits that come with it — preferential access to American chips, cloud infrastructure, and AI investment. The framing transforms Pax Silica from an economic-security association into something closer to a formal alliance structure, with membership dues paid in alignment.
Pax Silica: America's AI Alliance Project
Pax Silica was launched in December 2025 by Jacob Helberg, the State Department's Under Secretary for Economic Affairs, as the department's flagship effort on artificial intelligence and supply-chain security. According to State Department readouts, the initiative held its second summit in June 2026, brought on ten new partners, and plans a pilot AI supply-chain credentialing platform in Panama.
The June summit, where the AI Opportunity Statement was signed, marked the initiative's coming of age: 35 countries put their names to a document pledging cooperation on AI development and supply-chain security, and the Panama credentialing pilot began turning principles into infrastructure. Diplomatic readouts at the time described the statement as non-binding and inclusive. The new draft letter, if sent as written, would retroactively harden that soft commitment into an either-or choice — a bait-and-switch, critics are likely to argue, that risks alienating exactly the mid-aligned countries Washington most needs to keep.
The project has not been without friction abroad. In the Philippines, a proposed Pax Silica hub in New Clark City, Tarlac has drawn opposition from scientists, environmentalists, and farmers, who marched on the US Embassy in Manila in July, and the country's Department of Environment and Natural Resources has launched an environmental impact assessment of the project.
Why Now: The Open-Weight Squeeze
The drafting of the letter comes against a backdrop of mounting anxiety in Washington over Chinese AI momentum. As noted in coverage of the Reuters report, Chinese open-weight AI models have made rapid gains against proprietary systems from US companies, eroding the technical lead that American firms once treated as a strategic moat.
That anxiety is producing conflicting policy impulses. While the State Department drafts alliance-style commitments, more than 20 technology companies — including Nvidia, Microsoft, and Meta — are urging US policymakers not to broadly restrict open-weight AI models, arguing the technology is critical to American competitiveness. Meta CEO Mark Zuckerberg, meanwhile, has publicly opposed banning Chinese models, framing open ecosystems as the healthier long-term path.
A Broader Decoupling
The letter fits a wider pattern of technological separation. A separate Reuters exclusive on August 13 reported that Microsoft is retreating in China, even as the AI boom helps it keep a window open in the market. Washington has already moved against Chinese AI hardware in its supply chains, drafting restrictions on Chinese optical transceivers for data centers and banning Chinese humanoid robots on national security grounds. Each measure narrows the space where countries and companies can straddle both blocs.
Reuters has also reported that the United States and China plan to hold talks on artificial intelligence in September, according to people familiar with the matter — a diplomatic channel that will unfold alongside, rather than instead of, the coalition hardening. How smaller countries respond to the draft letter, once it is sent, may determine whether the AI economy splits into two incompatible stacks or remains a contested, connected whole.
What It Means for the AI Industry
For AI companies and cloud providers, the letter signals that market access may increasingly hinge on geopolitical alignment. Countries forced to choose between US and Chinese AI stacks could fragment global infrastructure, standards, and model distribution — raising costs for everyone and creating compliance minefields for vendors operating across blocs.
The draft letter has not yet been sent, and its final language could soften. But the direction is clear: after a year of building voluntary frameworks, Washington is moving toward demands, and the AI industry's global map is being redrawn along geopolitical lines.
---
Stay Ahead of AIGet the latest AI news, analysis, and breakthroughs — all in one place.
Read more AI news →
